arXiv:2607.01919cs.AIcs.CR2026-07

为智能体系统设计状态连续性协议,防止上下文被篡改。

ElephantAgent: Contextual State Continuity in Agentic Systems

论文配图:ElephantAgent: Contextual State Continuity in Agentic Systems
图 1 · 摘自论文原文
  • 通过重新计算并验证上下文摘要,确保状态连续性。
  • 利用可信硬件维护可线性化的状态变更记录。
  • 支持事后追溯与回滚,抵御恶意语义滥用。

智能体系统通过调用外部工具和持久化记忆来增强能力,但这些外部依赖引入了新的攻击面。近期的工具和记忆污染攻击表明,恶意构造的工具描述符和被污染的记忆可隐蔽地偏移代理行为。这反映出更深层问题:缺乏对规划与执行中上下文状态的可验证连续性。本文提出 ElephantAgent 协议,强制实施上下文状态连续性以防御上下文状态污染。受先前状态连续性机制(如 Nimble)启发,ElephantAgent 将该保护扩展至智能体系统的动态上下文状态。我们将上下文状态定义为代理完整上下文中受安全约束的有限子集(如工具状态和记忆)。在处理每个查询前,ElephantAgent 重新计算本地上下文状态的摘要,并与最新授权摘要进行比对。借助复制的可信硬件,ElephantAgent 维护一份可线性化的授权上下文状态转换日志,可检测带外状态篡改。为应对带内语义滥用,ElephantAgent 还提供历史可追溯性,支持条件性的事后审计与恢复至已知良好状态。

原文摘要 · Abstract (English)

Agentic systems enhance their capabilities by invoking external tools and maintaining persistent memory. However, these external dependencies introduce novel attack surfaces. Recent tool and memory poisoning attacks show that maliciously crafted tool descriptors and poisoned memory can covertly bias agent behavior. These threats reflect a deeper issue: the lack of verifiable continuity in the agent's contextual state for planning and execution. We present ElephantAgent, a protocol that enforces Contextual State Continuity to defend against contextual state poisoning. Inspired by prior state-continuity mechanisms (e.g., Nimble), ElephantAgent extends this protection to the evolving contextual state of agentic systems. We define the contextual state as the bounded, security-critical subset of the agent's entire context (e.g., tool state and memory). Before processing each query, ElephantAgent recomputes the digest of the local contextual state and verifies it against the latest authorized digest. Using replicated trusted hardware, ElephantAgent maintains a linearizable ledger of authorized contextual state transitions and detects out-of-band state tampering. To handle in-band semantic abuse, ElephantAgent additionally provides Historical Traceability, enabling conditional post-hoc audit and recovery to a known-good prior state.

智能体系统状态连续性安全防护可信硬件

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。